[bwilson4web]

Sorry, here is a cut-and-paste assembly:

Originally Posted by Jun 7, 2009

I can not rule out that E30 has similar fuel consumption rates as straight gas:

speed(mph), E30(MPG), expected(MPG), comment
61, 52.1, 53, Close
71, 44.0, 47, Close considering normal drop 70-75 mph is 49-39 MPG

I also ran a hill climb test:

speed(mph), E30(gal), Expected(gal), comment
55, .0738, .0720, Close

The problem is these differences can easily be explained by temperature
differences. If there is a difference, it is easily within the margin of error
for MFD based testing.

From a practical point of view, mixing E30 in the tank is a pain. There is a
small economic benefit:

E30 = E85@$2.15 + 2.75*E10@$2.35 (local prices)

I am not sure I will pursue this investigation until I improve my NHW11
instrumentation. The benefit appears to be marginal and we're looking at
fractional changes. I really want to find a metric that I can correlate to using
E30 versus straight gas, say spark advance. We do have a variable valve
mechanism but measuring it is not trivial.

I'm putting further investigations aside for now. I'm switching back to straight
gas since I'll need an accurate baseline before proceeding.

Originally Posted by Jun 7, 2009

. . ..
>
> From a practical point of view, mixing E30 in the tank is a pain. There is a
small economic benefit:
>
> E30 = E85@$2.15 + 2.75*E10@$2.35 (local prices)
>

My apologies. I'm fighting a cold (I hope) and didn't explain this well:

3.75*E30 = 1*E30 + 2.75*E10 # Mixture ratio if E10 and E30 are available

So doing the cost for prices a couple of weeks ago:

$2.15/gal = E85
$2.25/gal = E10
$2.22/gal = E30

So for saving $.02/gal, a 1% cost savings, have I recovered more than a 1%
improvement in performance OVER E10? But to make such a judgement call, I need
very accurate measurements over longer periods of time. Plus, I have to be
willing to put up with the dual fill-up and carry a small "topper" container.

A "topper" container is used to top off the tank after filling most of the tank
with fixed ratios of E10 and E30. A smaller, spare can, when it is empty, I fill
it with the fixed ratio and then complete the tank filling. But there is a
fundamental question: Is it worth it?

I want to know but in the meanwhile, I'm going back to straight gas. I'll need
to improve my instrumentation and methodology. In particular, I don't want to
have to carry fuel in the trunk. So I'm thinking of a running fuel log.

By accurately recording each fill-up of straight gas and keeping an accurate
fuel burn record, I can add 1 gallon of E85 when the fuel level is low enough to
make an E30 mix. The advantage is a 1 gallon (or a set of 1 gallon containers)
of E85 can be stored safely in a remote shed at our home. There will be a couple
of days during the burn-down while the straight gas dilutes the E30 until there
is enough space in the tank for 1 gallon of E85 to bring the mix back to E30.
However, I can use these intervals to track the vehicle fuel trim and spark
advance.

With these changes, I should be able get the accurate data needed to evaluate
both the fuel burn and economic impact. But these are small values.

Originally Posted by Jun 7, 2009

>
> We appreciate all the investigations you have pursued
> and information you have posted, but I wonder whether this
> test can yield any conclusive result. We expect a very
> small difference with a large number of variables
> involved. It would be difficult to even identify all
> the significant variables.

I gain much more from the peer review. Unlike other forums, I know a serious
posting is likely to have serious answers ... the occasional joke aside. But
seriously, the path to understanding is to share what is being done and using
the feedback to tweak the experiment.

> Some of these variables can be controlled or measured. Some
> cannot. We could probably make a large list including
> barometric pressure, wind, temperature of air, tires,
> transaxle, etc., but how to determine which can be ignored.

Using vehicle MFD, I agree. Fortunately, I have a Graham mini-scanner and many
of the metrics eliminate the problems of MFD metrics.

>
> It is one thing to experiment on a dynamometer in the lab,
> where fuels can be quickly switched while other parameters
> are nearly the same, but comparing road tests at different
> times can be very difficult. I suppose the initial question
> will be how repeatable are your results over a significant
> number of runs with straight gas. If the data looks promising,
> you could repeat the same test protocol with a different fuel
> and, with sufficient data, develop some probability estimates
> of the results. I'm not a statistician so can't help with
> the details.

I can record up to six metrics and my thinking is:

1) ICE rpm
2) injector timing
3) MG1 torque
4) spark advance
5) MG2 rpm
6) ICE temp

The first three allow tracking the brake specific fuel consumption, the amount
of fuel burned per engine shaft power. The spark advance is the most likely trim
function and MG2 provides speed. The ICE temperature helps eliminate cold engine
effects.

> It is an interesting experiment, but I agree with questioning
> "Is it worth it?"

I've seen E10, ethanol blamed for a number of MPG drops of different amounts.
But I don't think anyone has done a Prius specific study and this may be the
first. It will take time and it will be limited to the NHW11/NHW20 models.
However, it may give some insights.

Originally Posted by Jun 7, 2009

> OK, I'm too lazy to do the math, about how much E85, do I add to Gasolene with
10% methonal to get the E30?

1 part - E85
2.75 parts - E10

Originally Posted by Jun 7, 2009

> Would it not be simpler to just add straight metholated
> spirits (100% ethanol) to straight petrol and be sure you got
> the right figure. ? Or 47,5% rum. They say the exhaust
> smells better.

It is easier to use one of the traditional gasoline stations.

Bob Wilson